Sendspark

Security checks across malware telemetry and agentic risk

Overview

This Sendspark integration appears coherent, but it gives broad authenticated API and raw proxy power without clear confirmation rules for changes or deletion.

Install only if you are comfortable granting an agent access to your Sendspark account through Membrane. Prefer documented Membrane actions, and require explicit confirmation before creating, sending, updating, or deleting Sendspark data, especially when using raw proxy requests.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The skill description is broad ('Manage data, records, and automate workflows') and does not clearly constrain when the agent should invoke it or what kinds of Sendspark operations are appropriate without confirmation. In an agent setting, vague trigger boundaries can cause over-selection of the skill and unintended access or modification attempts against external data.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The proxy-request section explains how to send arbitrary HTTP methods, headers, and bodies to the Sendspark API but does not warn that these requests may create, modify, or delete data. That omission increases the chance an agent or user will treat the feature as routine and perform destructive operations without informed consent or adequate safeguards.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal